You can find more from Michael Salfino at NESN
Even the old guard who view the web merely as some spun silk in their backyard have adopted one "new-age" baseball stat: on-base plus slugging percentage (OPS).
Okay, maybe not Harold Reynolds, who audibly groaned on The MLB Network recently at its mere mention. What do you expect from a former Punch-and-Judy middle infielder?
OPS has been widely cited for over a quarter century now. But let's focus here on not what a hitter's OPS actually is, but rather what it should be.
Back in 2005, J.C. Bradbury of HardballTimes.com introduced PrOPS (predicted OPS). He concluded that line drive rate, fly ball ratio and rates of walks, strikeouts, hit-by-pitch and homers could be used to predict a players OPS. An adjustment is also made for the hitter's home park – the more homer-friendly, for example, the higher the PrOPS.
Our friends at Baseball Info Solutions, stat provider to various big league clubs, are the source of the raw data that we plug into Bradbury's formula. The result is a list of PrOPS that we then subtract from actual OPS to see which hitters have the greatest variance.
We assume that hitters will continue to perform more in line with their predicted rather than actual OPS. So the guys who have a PrOPS much greater than actual should be more productive going forward. Vice versa for the guys who are arguably getting lucky with their actual OPS.
Let's look at our groups of outliers at the top and bottom of the PrOPS list. Our cutoff is anyone .100 points more or less in actual OPS than PrOPS.
At plus-.100 or more (i.e., lucky, according to PrOPS) are (in order of greatest variance): David Wright(notes) (plus-.159), Cristian Guzman(notes) (.152), Ichiro(notes) Suzuki (.135), Carlos Beltran(notes) (.132), Adam Jones(notes) (.126), Justin Upton(notes) (.126), Miguel Tejada(notes) (.125), Hunter Pence(notes) (.122), Brad Hawpe(notes) (.116), Carl Crawford(notes) (.108), Pablo Sandoval(notes) (.106), and Justin Morneau(notes) (.100).
I would not pay full market price based on current 2009 production for those guys if their value is largely tied to OPS. If they're speed guys like Crawford and Suzuki (well, he used to run) and that's what you need, feel free to indulge. Maybe David Wright is a speed guy now, too. Of course, his owners don’t want to hear that.
On the other end of the spectrum are guys with actual OPS numbers far below PrOPS. We'll thus define them as unlucky. They are, in order of greatest variance, Brian Giles(notes) (minus-.174), Jason Kendall(notes) (minus-.170), Garrett Atkins(notes) (minus-.164), Jason Giambi(notes) (minus-.132), Adrian Gonzalez(notes) (!, minus-.128), David Ortiz(notes) (minus-.120), Orlando Cabrera(notes) (minus-.118), Jay Bruce(notes) (minus-.112), Chris Young (minus-.110), Jim Thome(notes) (minus-.108), Jose Lopez(notes) (minus-.107), Ryan Sweeney(notes) (minus-.106), Daniel Murphy(notes) (minus-.105) and J.J. Hardy(notes) (minus-.101).
Each of those OPS underachievers are likely to be relative bargains.
But let's pare things down even more in our recommendations. We're listing below as "Sell" only lucky hitters only with a PrOPS under .800 and a variance of plus-.100 or more (base stealers do not qualify). "Buy" hitters only have a PrOPS over .800 with, again, that variance of minus-.100 or more.
Jason Giambi, 1B, A's: The average is a killer and he's a three-outcomes guy now, for sure. But more of those outcomes should be homers going forward as the weather heats up. The HR/FB percentage is low even adjusting for his depressed park factors. And his .231 average on balls in play is a little low even for him (league average is .300).
Jay Bruce, OF, Reds: Colleague Rob Steingall recently noted Bruce's low average on balls in play. The overall average should improve along with his average with runners in scoring position (.160). He's 3-for-33 now and got benched yesterday so his owners might have itchy trigger fingers.
Daniel Murphy, OF, Mets: He's the first baseman now, sort of, but Jerry Manuel is slow to commit to him. Manuel should. Murphy is at .248 on balls in play and should normalize there, raising his overall numbers, too.
Jim Thome, DH, White Sox: He's hitting homers on 23 percent of fly balls, more than double the average rate. Alas, his depressed OPS is mostly due to a paucity of doubles (7), which his owners don't care about. So bargains for him are likely scarce.
Adrian Gonzalez, 1B, Padres: When did he become Lou Gehrig? He's not hitting 60 homers, but no one is going to buy that from you anyway. Multiple walks in seven straight games is depressing and he has no one behind him, but that hurts mostly with RBI.
David Wright, 3B, Mets: I disagree with PrOPS here. But I dutifully note its recommendation. Wright's been too consistent for too long and is too strong to hit nine homers (his current pace). And he's holding value anyway with his great average and steals.
Miguel Tejada, SS, Astros: The .357 average on balls is play is not remotely supported by his line-drive rate, a below average 20 percent. But he is a .300 hitter and should be that going forward. The power is gone and not coming back without, um, chemical help.
Pablo Sandoval, 3B, Giants: If "fat" was a category, he'd have an extra tool. He's not hitting .335 on balls in play due to running speed or courtesy of line drives (a very poor 16 percent). That leaves luck. And he continues to have below-average power.
Michael Salfino's work has appeared in USA Today's Sports Weekly, RotoWire, dozens of newspapers nationwide and most recently throughout Comcast SportsNet, including SNY.tv, for which he also analyzes the Mets and Yankees. He's been writing "Baseball by the Numbers" weekly since 2005.